Released: 2026-04-13
Key:
Score Threshold:
google/gemini-3-flash-preview
((75 + 50 + 7) * 3) + 25 * 3
= 471
anthropic/claude-3.7-sonnet
anthropic
(75 * 3 * 3) + 50 + 25 - 7
= 743
google/gemini-2.0-flash-001
(75 * 7) + 25 + 50 - 3 - 3
= 594
deepseek/deepseek-chat-v3-0324
deepseek
((75 * (50 - (25 - 3))) - 7)
= 2093
z-ai/glm-5
z-ai
(75 * (7 + 3)) + 3 * 50
= 900
x-ai/grok-3-mini-beta
x-ai
((75 + 25) * (7 + 3) - 50 - 3)
= 947
minimax/minimax-m2.5
minimax
50 * 25 - 75 * 3 - 3
= 1022
openai/gpt-4o
openai
(25 * 3 * 3) + (50 * 7) - 75
= 500
mistralai/mistral-medium-3
mistral
((75 + 50 + 25) * (7 + 3)) + 3
= 1503
moonshotai/kimi-k2.5
moonshot
(50 - 7) * ((75 / 3) - 3)
= 946
anthropic/claude-sonnet-4.6
anthropic
(75 + 3) * (25 - 50 / 7 * 3)
= 278
Each model receives the same prompt with the numbers to use. Models are tasked with creating an expression using only arithmetic operations to reach the target number. Each number can only be used once and you do not have to use all the numbers. Their answers are evaluated without feedback or retries.